Circulation Journal
Online ISSN : 1347-4820
Print ISSN : 1346-9843
ISSN-L : 1346-9843
Imaging
Anatomical and Functional Discrepancy in Diabetic Patients With Intermediate Coronary Lesions ― An Intravascular Ultrasound and Quantitative Flow Ratio Study ―
Liang GengXibao ShiYuan YuanPeizhao DuLiming GaoYunkai WangJiming LiWei GuoYing HuangQi Zhang
著者情報
ジャーナル オープンアクセス HTML

2023 年 87 巻 2 号 p. 320-328

詳細
Abstract

Background: Data regarding the performance of computational fractional flow reserve in patients with diabetes mellitus (DM) remain scarce. This study sought to explore the impact of DM on quantitative flow ratio (QFR) and its association with intravascular ultrasound (IVUS)-derived anatomical references.

Methods and Results: IVUS and QFR were retrospectively analyzed in 237 non-diabetic and 93 diabetic patients with 250 and 102 intermediate lesions, respectively. Diabetics were further categorized based on adequate (HbA1c <7.0%: 47 patients with 53 lesions) or poor (HbA1c ≥7.0%: 46 patients with 49 lesions) glycemic control. Lesions with QFR ≤0.8 or minimum lumen area (MLA) ≤4.0 mm2 and plaque burden (PB, %) ≥70 were considered functionally or anatomically significant, respectively. PB increased, and MLA decreased stepwise across non-diabetics, diabetics with adequate glycemic control and those with poor glycemic control. In contrast, QFR was similar among the 3 groups. PB correlated significantly with the QFR for lesions in non-diabetics, but not for lesions in diabetics. DM was independently correlated with the functionally non-significant lesions (QFR >0.8) with high-risk IVUS features (MLA ≤4.0 mm2 and PB ≥70; OR 2.053, 95% CI: 1.137–3.707, P=0.017). When considering the effect of glycemic control, HbA1c was an independent predictor of anatomical-functional discordance (OR 1.347, 95% CI: 1.089–1.667, P=0.006).

Conclusions: Anatomical-functional discordance of intermediate coronary lesions assessed by IVUS and QFR is exacerbated in patients with diabetes, especially when glycemia is poorly controlled.

Numerous clinical studies have shown that assessment of functional significance of coronary lesions by fractional flow reserve (FFR) plays a pivotal role in the guidance of percutaneous coronary intervention (PCI) strategy and evaluation of long-term outcomes.1,2 The rationale of physiology-based revascularization has been further expanded by the non-inferiority of instantaneous wave-free ratio (iFR) for assessing functional significance of coronary lesions in comparison with FFR.3,4 More recently, the FAVOR III China trial demonstrated that PCI guided by quantitative flow ratio (QFR) provides superior prognostic implication over plain angiography guidance.5

Although the skill set repertoire of interventional cardiologists to identify the functionally significance of intermediate lesions has been largely expanded, concerns regarding the safety of deferred lesions with negative FFR are increasing, especially for certain clinical scenarios such as diabetes mellitus (DM).68 On one hand, reliability concerns largely exist as diffused lesion, impaired hyperemic response and microvascular dysfunction may attenuate FFR measurement in DM patients.9 In contrast, initial validation studies using plain angiography-derived anatomical severity as the reference did not show any difference regarding the diagnostic performance of FFR between diabetic and non-diabetic populations.10,11 Nevertheless, 2-dimensional quantitative coronary angiography has inherent limitations in evaluating plaque burden (PB) or its morphology, particularly when patients are complicated with diabetic status.12 Likewise, as a virtual surrogate of FFR, the diagnostic performance of QFR is well studied among the general population, but has been less validated in diabetic patients. Therefore, we sought to determine the impact of DM patients with or without adequate glycemic control on the QFR-defined functional significance of intermediate coronary lesions, using intravascular ultrasound (IVUS)-derived anatomical severity as the reference.

Methods

The study protocol was carried out in accordance with the Declaration of Helsinki and was approved by the institutional review board of Shanghai East Hospital. Individual consent for this retrospective analysis was waived.

Study Population

We performed a retrospective analysis of a total of 448 coronary artery disease patients with 482 intermediate lesions (visual estimation of diameter stenosis [DS] of 50–75%) recruited from an institutional IVUS registry at Shanghai East Hospital (Shanghai, China) between January 2015 and December 2018, as previously described.13 The diagnosis of DM was made based on the criteria of the American Diabetes Association.14 For the purpose of the current study, complex lesions or lesions without adequate imaging qualities for IVUS or QFR measurements were excluded. Therefore, 330 patients with 352 intermediated lesions were included in the final analysis. Of them, there were 237 non-diabetic (250 lesions) and 93 diabetic (102 lesions) patients. All diabetic patients were further categorized based on adequate (HbA1c <7.0%; n=47, 53 lesions) or poor (HbA1c ≥7.0%; n=46, 49 lesions) glycemic control (Figure 1).

Figure 1.

Study flowchart. DM, diabetes mellitus; IVUS, intravascular ultrasound; L, lesion number; N, patient number; QFR, quantitative flow ratio.

IVUS Measurement

IVUS imaging acquisition and assessments were performed using the standard protocol described previously.15,16 Briefly, a 40-MHz IVUS catheter was advanced into the 10 mm distal reference site of the target lesion, and an automatic pullback was performed using the iLab system (Boston Scientific, Natick, MA, USA). IVUS runs were recorded and off-line analyzed by 2 experienced physicians who are blinded to QFR data. External elastic membrane (EEM) cross-sectional area (CSA), lumen CSA, and plaque plus media (P+M) CSA were automatically traced at the lesion site and 5-mm proximal and distal reference sites using validated planimetry software (Image viewer, Boston Scientific). The lesion site was the site with the minimum lumen area (MLA), whereas the reference site was the site with the largest lumen and smallest PB without any intervening side branch. PB (%), defined as the (P+M) CSA divided by the EEM CSA, was automatically calculated in the planimetry software. An anatomically significant lesion was defined according to the classical IVUS cut-offs of MLA ≤4.0 mm2 and PB ≥70. Other IVUS-derived anatomical parameters including area stenosis (AS, %), remodeling index (RI) and plaque eccentricity were calculated according to the published standards.17

QFR Determination

QFR was determined off-line by 2 qualified interventional cardiologists who were blinded to the IVUS data, using the commercially available QFR system, AngioPlus (Pulse Medical Imaging Technology, Shanghai Co., Ltd., Shanghai, China). Three-dimensional reconstruction was accomplished by automatically contouring the lumen of the interrogated vessel.15,18 Contrast flow model using the frame count method was introduced to derive the final QFR value. The 3-dimensional quantitative coronary angiography (3D-QCA) parameters were derived from the QFR report. QFR ≤0.8 was considered as functionally significant.

Calculation of QFR-Derived Indices

The function patterns of vessels were evaluated using the QFR-derived hyperemic pullback pressure gradient (PPG) index, as previously described.1921 Pullback curves from QFR diagrams were extracted and digitalized using the OriginPro, Version 2021 (OriginLab Corporation, Northampton, MA, USA). PPG index was calculated as follows:

where maximal PPG20 mm (MaxPPG20 mm) was defined as the maximum pressure drop over 20 mm, and ∆QFRvessel as the difference between QFR values at the ostium of the vessel and at the distal end of the vessel. The length of functional disease was defined as the length with a QFR drop ≥0.0015/mm.22 A high PPG index (approaching 1.0) indicates focal disease, whereas a low PPG index (close to 0) indicates diffuse disease.22

Angiography-derived index of microvascular resistance (IMRangio) was calculated according to the method described previously and used to assess coronary microvascular status.15,23

where Pa is baseline aortic pressure; Nframes is the number of frames for contrast dye running from the tip of the catheter to the distal end of the interrogated vessel; and fps is the frame rate during the acquisition (prespecified as 15 frames per second).

Statistical Analysis

Continuous variables are expressed as mean±standard deviation and compared among groups using the one-way ANOVA. Categorical variables are expressed as frequencies (percentages) and compared using a chi-squared test. Relationships between QFR and IVUS-derived parameters were assessed by using Pearson correlation analysis. The correlation coefficients were compared via Fisher’s z-transformation. Multiple logistic regression analysis was used to identify the independent predictors of discordance between QFR and IVUS parameters. Variables in univariable analysis (P<0.1) and potentially clinical confounders were included in the multivariable model. The multivariable adjustment was performed with a backward selection method employing a threshold P>0.15 for removal. A statistical significance was defined when 2-sided P<0.05. The statistical analyses were performed with SPSS version 22.0 (SPSS Institute, Chicago, IL, USA) and GraphPad prism 7.00 (GraphPad Software, Inc, La Jolla, CA, USA).

Results

Clinical Characteristics

There were no significant differences in demographics, clinical characteristics, and laboratory findings among the 3 groups except for older age and higher HbA1c levels in diabetic cohorts. In addition, the percentage of patients with insulin dependence was higher in diabetic patients with poor glycemic control (Table 1).

Table 1. Clinical Characteristics Stratified by Diabetic Status
  Non-DM
(n=237)
DM with HbA1c
<7% (n=47)
DM with HbA1c
≥7% (n=46)
P value
Age, years 64.4±9.4 69.0±7.1 66.1±10.3 0.006
Male 143 (60.3) 24 (51.1) 26 (56.5) 0.479
Hypertension 154 (65.0) 36 (76.6) 32 (69.6) 0.282
Hypercholesterolemia 12 (5.1) 1 (2.1) 3 (6.5) 0.589
Current smoker 82 (34.6) 18 (38.3) 16 (34.8) 0.888
Clinical presentation       0.150
 Acute myocardial infarction 4 (1.7) 0 2 (4.3)  
 Unstable angina 5 (2.1) 2 (4.3) 2 (4.3)  
 Atypical angina 87 (36.7) 20 (42.6) 22 (47.8)  
 Stable angina 124 (52.3) 19 (40.4) 14 (30.4)  
 Silence ischemia 17 (7.2) 6 (12.8) 6 (13.0)  
Multi-vessel disease 132 (55.7) 28 (59.6) 31 (67.4) 0.345
Prior PCI history 28 (11.8) 8 (17.0) 11 (23.9) 0.084
Prior CABG history 1 (0.4) 0 0 0.821
Prior MI 7 (3.0) 2 (4.3) 4 (8.7) 0.185
EF, % 62.8±3.8 62.3±5.1 62.6±4.3 0.768
BMI 24.2±4.1 24.0±3.3 24.4±3.1 0.946
Laboratory
 Creatinine, μmol/L 73.1±17.4 73.7±20.8 72.2±19.5 0.928
 LDL-C, mmol/L 2.6±0.9 2.3±0.9 2.5±1.0 0.095
 TG, mmol/L 1.5±0.8 1.4±0.7 1.7±0.9 0.122
 HbA1c, % 5.8±0.6 6.1±1.5 8.6±1.5 <0.001
Medications
 Aspirin 191 (80.6) 36 (76.6) 39 (84.8) 0.604
 Clopidogrel 90 (38.0) 20 (42.6) 15 (32.6) <0.001
 Statin 234 (98.7) 47 (100) 44 (95.7) 0.100
 β-blocker 138 (58.2) 24 (51.1) 35 (76.1) 0.034
 ACEI/ARB 102 (43.0) 24 (51.1) 24 (52.2) 0.386
 Insulin 0 3 (6.4) 15 (32.6) <0.001

Data are presented as mean±SD or n (%). ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; CABG, coronary artery bypass graft; DM, diabetes mellitus; EF, left ventricular eject fraction; HbA1c, glycated hemoglobin A1c; LDL-C, low-density lipoprotein cholesterol; MI, myocardial infarction; PCI, percutaneous coronary intervention; TG, triglyceride.

Lesion Characteristics

Overall, QCA-derived minimum lumen diameter decreased (1.9±0.5 mm vs. 1.8±0.3 mm vs. 1.7±0.4 mm, P=0.017), whereas IVUS-derived PB (65.0±9.1 vs. 67.6±7.7 vs. 69.6±8.5, P=0.002) increased stepwise across non-diabetics, diabetic patients with adequate glycemic control and those with poor glycemic control. There was a trend towards lowest IVUS-derived MLA in diabetic patients with HbA1c ≥7% (4.4±1.8 mm2 vs. 4.1±1.2 mm2 vs. 3.8±1.5 mm2, P=0.074). In contrast, the physiological indices, including fix-flow QFR (fQFR), contrast-flow QFR (cQFR), PPG index, resting flow velocity and IMRangio, were similar among the 3 groups (Table 2).

Table 2. Lesion Characteristics Stratified by Diabetic Status
  Non-DM
(L=250)
DM with HbA1c
<7% (L=53)
DM with HbA1c
≥7% (L=49)
P value
Lesions with QFR ≤0.8 40 (16.0) 5 (9.4) 9 (18.4) 0.396
3D QCA parameters
 LAD lesions 183 (73.2) 38 (71.7) 36 (73.5) 0.972
 Proximal lesions 129 (51.6) 27 (50.9) 20 (40.8) 0.381
 RVD, mm 3.2±0.5 3.1±0.5 3.2±1.0 0.662
 DS, % 40.3±9.0 41.0±6.1 42.6±7.4 0.197
 MLD, mm 1.9±0.5 1.8±0.3 1.7±0.4 0.017
 Lesion length, mm 24.8±13.2 23.8±12.2 24.2±11.3 0.853
Physiological indices
 fQFR 0.88±0.09 0.88±0.07 0.86±0.09 0.426
 cQFR 0.88±0.09 0.89±0.06 0.87±0.08 0.478
 PPG index 0.84±0.12 0.83±0.10 0.81±0.11 0.256
 Resting flow velocity, m/s 0.17±0.10 0.18±0.11 0.17±0.07 0.862
 IMRangio 56.3±28.2 55.3±25.6 51.6±21.5 0.540
IVUS parameters
 MLA, mm2 4.4±1.8 4.1±1.2 3.8±1.5 0.074
 AS, % 58.2±11.7 59.3±11.2 61.3±10.3 0.200
 PB, % 65.0±9.1 67.6±7.7 69.6±8.5 0.002
 Eccentricity 0.76±0.17 0.74±0.18 0.79±0.17 0.396
 RI 0.86±0.21 0.88±0.19 0.85±0.18 0.744

Data are presented as mean±SD or n (%). AS, area stenosis; DS, diameter stenosis; DM, diabetes mellitus; cQFR, contrast-flow QFR; fQFR, fix-flow QFR; HbA1c, glycated hemoglobin A1c; IMRangio, angiography-derived index of microvascular resistance; IVUS, intravascular ultrasound; LAD, left anterior descending; L, leison; MLA, minimum lumen area; MLD, minimum lumen diameter; PB, plaque burden; PPG, pullback pressure gradient; 3D QCA, 3-dimensional quantitative coronary angiography; QFR, quantitative flow ratio; RI, remodeling index; RVD, reference vessel diameter.

Association of DM With the Discrepancy Between IVUS and QFR

For lesions with a QFR >0.8, the IVUS-derived anatomical severity, as well as the rate of IVUS-positive lesions (MLA ≤4.0 mm2 and PB ≥70%) increased stepwise across non-diabetics, diabetic patients with adequate glycemic control and those with poor glycemic control (Table 3). Likewise, the incidence of DM was significantly higher in QFR-negative but IVUS-positive lesions (41.7% vs. 26.4%, P=0.017) (Figure 2A). Figure 2B shows the combined QFR and IVUS measurements of an intermediate lesion in a diabetic patient.

Table 3. Lesion Characteristics of Non-Flow-Limiting Lesions (QFR >0.8) Stratified by Diabetic Status
  Non-DM
(L=210)
DM with HbA1c
<7% (L=48)
DM with HbA1c
≥7% (L=40)
P value
3D QCA parameters
 LAD lesions 146 (69.5) 34 (70.8) 37 (67.5) 0.944
 Proximal lesions 106 (50.5) 24 (50.0) 16 (40.0) 0.473
 RVD, mm 3.2±0.5 3.1±0.5 3.3±1.1 0.244
 DS, % 38.4±7.7 40.2±5.6 40.7±6.1 0.077
 MLD, mm 2.0±0.5 1.8±0.3 1.8±0.4 0.015
 Lesion length, mm 23.4±12.4 23.1±12.2 23.0±11.0 0.979
Physiological indices
 fQFR 0.90±0.06 0.89±0.06 0.89±0.06 0.301
 cQFR 0.92±0.05 0.90±0.05 0.90±0.05 0.132
 PPG index 0.86±0.11 0.84±0.10 0.83±0.11 0.154
 Resting flow velocity, m/s 0.16±0.09 0.17±0.11 0.17±0.06 0.676
 IMRangio 59.6±28.1 57.5±25.9 53.6±21.0 0.427
IVUS parameters
 MLA, mm2 4.6±1.8 4.1±1.2 4.1±1.5 0.086
 AS, % 56.8±11.6 58.9±11.2 60.6±9.4 0.115
 PB, % 64.1±8.8 67.4±8.0 68.9±8.4 0.001
 Eccentricity 0.76±0.17 0.73±0.18 0.77±0.18 0.370
 RI 0.86±0.20 0.89±0.20 0.86±0.19 0.715
Lesions with MLA ≤4.0 mm2 and
PB ≥70%
35 (16.7) 12 (25.0) 13 (32.5) 0.048

Data are presented as mean±SD or n (%). Abbreviations as in Table 2.

Figure 2.

(A) Incidences of DM stratified by the functional severity and anatomical severity. (B) Representative illustration of an intermediate lesion in a diabetic patient with anatomical-functional discordance. DM, diabetes mellitus; MLA, minimum lumen area; PB, plaque burden; QFR, quantitative flow ratio.

The correlation between IVUS-derived anatomical parameters and QFR is shown in Figure 3. Interestingly, PB correlated significantly with QFR in non-diabetic (r=−0.356, P<0.001) but not in diabetic patients (r=−0.150, P=0.134). The correlation coefficients between MLA and QFR were similar between non-diabetic and diabetic patients (r=0.373 vs. r=0.313, P=0.682).

Figure 3.

Correlations between IVUS-derived anatomical parameters and QFR in a DM and non-DM cohort. DM, diabetes mellitus; IVUS, intravascular ultrasound; MLA, minimum lumen area; PB, plaque burden; QFR, quantitative flow ratio.

Two models were constructed to predict the discrepancy between QFR and IVUS measurements (Table 4). In model 1, diabetes (OR 2.053, 95% CI: 1.137–3.707, P=0.017) was an independent predictor of QFR-negative (QFR >0.8) and IVUS-positive (MLA ≤4.0 mm2 and PB ≥70%) lesions. When taking into account the glycemic control (model 2), higher HbA1c level (OR 1.347, 95% CI: 1.089–1.667, P=0.006) and lower plaque eccentricity (OR: 0.818, 95% CI: 0.686–0.976; P=0.026) were selected as independent predictors of lesions with anatomical-functional discordance.

Table 4. Predictors of QFR-Negative (QFR >0.8) and IVUS-Positive (MLA ≤4.0 mm2 and PB ≥70%) Lesions
Variables OR (95% CI) P value
Model 1
 Diabetes 2.053 (1.137–3.707) 0.017
 Eccentricity, per 0.1 increase 0.875 (0.746–1.025) 0.098
 Multi-vessel disease 1.730 (0.920–3.253) 0.089
Model 2
 HbA1c, per 1% increase 1.347 (1.089–1.667) 0.006
 Eccentricity, per 0.1 increase 0.818 (0.686–0.976) 0.026
 Proximal lesion 0.543 (0.288–1.022) 0.058
 Multi-vessel disease 1.626 (0.834–3.172) 0.154
Univariate analysis
 Male 1.123 (0.637–1.978) 0.689
 Age 1.013 (0.982–1.044) 0.419
 Acute coronary syndrome 1.130 (0.312–4.092) 0.853
 Hypertension 0.709 (0.381–1.320) 0.278
 Hypercholesterolemia 0.637 (0.142–2.860) 0.556
 Diabetes 1.994 (1.122–3.546) 0.019
 HbA1c, per 1% increase 1.309 (1.071–1.599) 0.008
 Multi-vessel disease 1.826 (0.994–3.353) 0.052
 Proximal lesion 0.851 (0.488–1.485) 0.571
 LAD lesion 0.922 (0.497–1.711) 0.797
 Eccentricity, per 0.1 increase 0.880 (0.754–1.027) 0.105
 Lesion length, per 1 mm increase 1.007 (0.986–1.029) 0.495
 Diameter stenosis, per 1% increase 1.017 (0.983–1.051) 0.337

CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,2.

Discussion

Our results showed that in patients with DM, especially with poor glycemic control, coronary anatomical changes are often not translated into QFR-defined severity of physiological significance, leading to an anatomical-functional discordance in the assessment of intermediate coronary lesions.

It is well recognized that diabetes accelerates development of atherosclerosis and progression of macro- and micro-vascular diseases through multiple biochemical and molecular pathways.24,25 Post-hoc analysis of the DEFINE-FLAIR trial (Functional Lesion Assessment of Intermediate Stenosis to Guide Revascularisation) showed that diabetes is associated with increased risk of major adverse cardiac events in both iFR-guided and FFR-guided groups, and real-world registry studies have demonstrated a higher incidence of FFR-guided deferred lesion failure in patients with DM.7,26,27 The underlying mechanism that DM invalidates the prognostic capabilities of FFR remains elusive, as the intravascular imaging and microcirculation assessment were not routinely performed in previous studies.7,27

The major finding of our study is that DM is associated with discordance between IVUS-derived anatomical severity and QFR-defined functional severity in the assessment of intermediate lesions. In this study, IVUS-derived PB increased, and MLA decreased stepwise from non-DM to DM patients with poor glucose control, supporting the hypothesis that DM aggravates coronary atherosclerotic burden. However, the QFR value was similar in non-diabetic and diabetic patients, irrespective of glycemic status. Additionally, the relationship between IVUS-derived PB and QFR was significant only for the non-DM cohort. Interestingly, diabetic patients, particularly those with poor glycemic control, were more likely to have intermediate coronary lesions with abnormal IVUS features but a negative QFR value (QFR >0.8). These findings are in line with previous reports about invasive FFR determination, showing that FFR correlated poorly with optical coherence tomography- and QCA-derived anatomical parameters in diabetic patients.28,29

The mechanism of anatomical and functional discrepancy of intermediate coronary lesions in diabetic patients is multifactorial, but may be likely related to microvascular disease and diffuse coronary lesions.15,30,31 It has been shown that microcirculatory dysfunction interferes with the measurement of hyperemic indices, including FFR or virtual FFR, such as QFR.3234 In contrast, iFR, as a resting physiological index, is less affected by microvascular dysfunction, and there was a disagreement between iFR and FFR value in 15–20% of diabetic patients, and diabetes was an independent predictor of discordance between (high)FFR and (low)iFR.3537 In the DEFINE-FLAIR trial, diabetes-driven more advanced atherosclerotic burden was clearly reflected by the lower iFR value, whereas FFR was similar in the diabetic and non-diabetic groups.27 IMRangio has been proposed for physiological assessment of microvascular diseases in coronary circulation.23 In this study, IMRangio did not significantly differ among the 3 groups, which may partially be attributed to the contrast-induced submaximal hyperemia and bimodal pattern of microvascular involvement (decreased baseline microvascular resistance and increased hyperemic microvascular resistance) in DM patients.38,39 In addition, the possible influence of diffuseness of coronary atherosclerosis on anatomical and functional discordance in diabetes was evaluated using the QFR-derived PPG index, which has been shown to discriminate focal from diffuse coronary disease.1921 Intriguingly, the longitudinal lesion complexity determined by the QFR-derived PPG index was equally distributed among the 3 groups. The exact reason remains unclear, but may be, at least partly, explained by the exclusion of complex coronary lesions (e.g., tandem lesions), which may obscure the impact of DM on the physiological pattern of coronary disease. Nevertheless, further prospective studies are warranted to examine whether diabetes breaks the mechanistic link between anatomical and functional severity of intermediate lesions through accelerating the atherosclerotic process and microvascular disease in coronary circulation.

Previous studies have indicated that IVUS-derived MLA and PB are prognostic determinants of non-flow-limiting lesions.40,41 Recently, Cho et al found that DM and PB ≥70% were associated with long-term risks of FFR-guided deferred lesions.26 Theoretically, QFR-guided deferred intermediate lesions in diabetic patients possess higher residual IVUS-defined anatomical significance, which may lead to an excess of cardiovascular events, especially in those with poor glycemic control. Our findings highlight that intravascular imaging may further delineate the risk of intermediate lesions in diabetic patients, particularly in those with poor glucose control or a borderline QFR/FFR value (0.8–0.85).6 Future clinical trials are warranted to explore the clinical implication of QFR with a specific focus on diabetic cohort.

Study Limitations

We recognize limitations in our study. First, owing to the retrospective nature of this single-center study, there existed selection bias. Second, IVUS-derived MLA and PB were measured in certain selected slices, which may not reflect longitudinal information of the lesions; however, the longitudinal lesion complexity determined by PPG index was equally distributed among the 3 groups. Third, the accuracy and feasibility of computations of QFR and its derived indices were limited because of the retrospective nature. Fourth, we focused on the single and intermediate lesions and thereby conclusions should be cautiously drawn about mild and tandem lesions. Finally, this study is not outcome-oriented, thus prospective studies incorporating both intravascular imaging and QFR are warranted to further explore the natural history of deferred non-flow-limiting lesions in patients with DM.

Conclusions

In DM patients, especially for those with poor glucose control, intermediate lesions have a higher prevalence of anatomical-functional discordance between IVUS and QFR assessments.

Acknowledgment

The authors appreciate Dr. Wei Feng Shen for assistance with the preparation of the manuscript.

Sources of Funding

The research received grants from Top-level Clinical Discipline Project of Shanghai Pudong District (PWYgf2021-01), Leader Training Plan of Shanghai Pudong District (PWRd2018-06), Shanghai Key clinical specialty Project (shslczdzk06202), and Jiangxi Municipal Health Commission (202120110).

Disclosures

The authors have no conflicts of interest to declare.

IRB Information

The study protocol was in accordance with the Declaration of Helsinki and was approved by the institutional review board of Shanghai East Hospital (research project number: 2020/No.007).

References
 
© 2023, THE JAPANESE CIRCULATION SOCIETY

This article is licensed under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license.
https://creativecommons.org/licenses/by-nc-nd/4.0/
feedback
Top